Modelos de Classificação de Aprendizagem de Máquina para Priorização de Teste de COVID-19 no Brasil

• Íris Viana dos S. Santana UFAPE
• Álvaro Alvares de Carvalho César Sobrinho UFAPE

DOI: https://doi.org/10.5753/sbcas.2021.16111

Abstract

The aim of this study is to effectively prioritize symptomatic patients for testing COVID-19 in Brazil and thus assist in early detection, addressing problems related to testing and control strategies. 55,676 data were pre-processed, and the chi-square test was performed to confirm the relevance of fever, sore throat, dyspnea, cough, coryza, headache, olfactory and taste disorders. Classification models were implemented relying on data sets; supervised learning; and classic algorithms . One of the models with the highest performance was the decision tree (mean accuracy ≥ 89.12%), as it is easy to interpret, it was considered the most adequate.

References

Chatterjee, A., Gerdes, M. W., Martinez, S. G. (2020). Identification of risk factors associated with obesity and overweight-a machine learning overview. Sensors (Basel), 20(9):2734.

Malta, M., Murray, L., Silva, C. M. F. P., Strathdee, S. A. (2020). Coronavirus in Brazil: the heavy weight of inequality and unsound leadership. EClinicalMedicine, 25:100472.

Monteiro, O. M., Fuller, T. L., Brasil, P., Gabaglia, C. R., Nielsen-Saines, K. (2020). Controlling the COVID-19 pandemic in Brazil: a challenge of continental proportions. Nature Medicine, 26(10):1505-1506.

Santana Í. S. S., Silveira, A. C. M., Sobrinho, Á., Silva, L. C., Silva, L. D., Santos, D. F. S., Gurjão, E. C., Perkusich, A. (2021). A Brazilian dataset of symptomatic patients for screening the risk of COVID-19. Mendeley Data.

Santana, Í. V. S., Silveira, A. C. M., Sobrinho, Á., Silva, L. C., Silva, L. D., Santos, D. F. S., Gurjão, E. C., Perkusich, A. (2021). Classification Models for COVID-19 Test Prioritization in Brazil: Machine Learning Approach. Journal of Medical Internet Research, 23(4).